European Journal of Molecular & Clinical Medicine ISSN 2515-8260 Volume 07, Issue 09, 2020
2881
PRIORITY BASED CACHE
ALLOCATION AND CONGESTION
CONTROL PROTOCOL IN WIRELESS
SENSOR NETWORK
Ms. K. Manjula1, Dr. P.V.S. Siva Prasad
2*, Mr. N. Srinivas
2
1. M.Tech, Nalla Malla Reddy Engineering College, Hyderabad.
Email: [email protected],
2*. Professor, Ph.D(CSE), Ing.Pead.IGIP, Nalla Malla Reddy Engineering College,
Hyderabad.
Email: [email protected]
2. Associate Professor, Vignana Bharathi institute of technology , Aushapur , Ghatkesar,
Hyderabad.Email:[email protected],
Abstract: Reliable data transmission in wireless sensor network plays a critical part with the
presence of increased data traffic and congestion. Successful data transmission can be
ensured by providing cache resources to store the data before transmission. Reliable Transport
protocol with a cache aware congestion control mechanism (RT-CaCC) is the existing
approach used for secured communication. However, in this existing works, cache resource
allocation is done evenly to all data flows on network which lead to data transmission failure
due to insufficient resources. Cache resource allocation should be performed evenly in order
to obtain consistent and successful data transmission. Priority based Cache Allocation and
Congestion Control protocol (P-CACC) is used in this work for effective cache resource
allocation. In the proposed research work, data transmission is performed reliably by giving
more prioritization to the data flow with higher data rate. Here, initially uneven cache
partitioning scheme is introduced for giving more prioritization to the flow with higher data
rates. This is done by measuring the priority of the multiple data flows present in the network,
and the weight assignment is done with the assistance of cuckoo search algorithm where the
data flow with more priority will be assigned with more weight. The higher cache partition will
be allocated to the data flow assigned with higher weight value. The simulation of the work is
performed in NS2. It is confirmed that proposed research work tends to have improved
outcomes than the existing techniques.
Keywords: Priority, Caching Resources, Weight Assignment, Cache Partition, Higher Data
Rate, Cuckoo Search Algorithm.
European Journal of Molecular & Clinical Medicine ISSN 2515-8260 Volume 07, Issue 09, 2020
2882
I. INTRODUCTION
In modern days, developments in low-power wireless communications as well as in
microelectronics have made possible to short-size and less-cost microsensors growth, that
communicate along with each other through radios [1]. A huge quantity of wireless microsensors
generate Wireless Sensor Network (WSN), in that sensor nodes cooperate for sensing actions or
phenomena of significance, process sensed data and transmit them to customers of information
(data sink) [2]. In the 21st century, WSNs are noticed as one of the very most valuable
technology. Nevertheless, numerous technical confronts need to be concentrated before extensive
business deployment of WSNs turned out to be realistic. Here, one of these confronts is, data
transport over WSNs [3]. The transport layer protocols that are broadly-utilized in Internet,
namely, User Datagram Protocol (UDP) and Transmission Control Protocol (TCP) possibly will
not be appropriate for data transport in WSNs because of exact features of WSNs and novel
necessities of applications over WSNs [4].
Wireless sensor nodes dispersed and energy-restricted are routinely processed with the
cooperative monitoring or control of one or a few functions in a WSN[5]. The pre-processing or
aggregation of event data in the intermediate sensor nodes is feasible and appropriate prior to
additional transmissions in the direction of the data sinks for which the sensor data is assembled
and the sensor nodes are managed[6].While, conventional transport layers have not been
intended along with some consideration of these novel characteristics of WSNs, there is a
requirement for improved transport protocols [7]. There are, two main tasks in transport layer,
which are reliable data delivery and congestion control [8]. It needs that when packets are
misplaced in a multihop WSN, a few or entire of misplaced packets is identified and misplaced
information gathered through suitable mechanisms [9]. Reliability is needed during information
is significant to a function in WSNs.
Congestion happens when numerous sensor nodes transmit data to sinks and data jamming
quantity goes beyond network capability [10]. However, during congestion, nodes begin to drop
packets or packets delay considerably goes beyond applications necessity. Repeated dropping of
packets is a dissipation of energy and neutralize some attempt to attain consistency. Extreme
delay of packets possibly will produce invalidity of data gathered by sensors [11]. Congestion
controls intends to neglect congestion or simplify congestion intensity. Attempts have been
carried out in recent times to manage consistent transport and congestion control issues in
WSNs, such as, Event to Sink Reliable Transport (ESRT), Reliable Multisegment Transport
(RMST), Congestion Detection and Avoidance (CODA), Pump Slowly Fetch Quickly (PSFQ),
and GARUDA aiming consistent data transport or congestion control or both.
Here, reliable and congestion aware data transmission is guaranteed through allocating the
cache resources to the various data flows in accordance with requirements. Here uneven resource
allocation is guaranteed by measuring the priority of the data flows happening on the network.
This is carried out by measuring the priority level of various data flows and higher cache
resources are allocated to the data flow with increased priority level.
European Journal of Molecular & Clinical Medicine ISSN 2515-8260 Volume 07, Issue 09, 2020
2883
The organization of this research work is provided here. In this section 1, described
introduction regarding the necessity of congestion and reliability consideration at the time of
executing the data transmission in the WSN. In section 2, investigation of several relevant
research methods that are commenced to realize the congestion avoided consistent data
transmission. In section 3, proposed research method had been described thoroughly together
with the appropriate figures and examples. In section 4, performance analysis of the proposed
research method has been provided in accordance with numerical assessment. In the end section
5, on the whole, ending of the research work is provided in terms of getting experimental result.
II. RELATED WORK
A new bio-inspired routing protocol which is termed as CB-RACO is implemented by Rosset
et al [12]. In this protocol, a distributed community finding as well as computationally
inexpensive technique called Label Propagation (LP) is combined with Ant Colony Optimization
(ACO) meta-heuristic algorithm. In WSNs, communities are generated using CB-RACO and
through routing data inside-communities, it gathers energy utilization via swarm intelligence. It
requires short memory, but need routing path safeguarding and enhances structure overhead.
For WSNs, a TDMA-based reliable multicast MAC (TRM-MAC) protocol is implemented by
Bhatia et al [13]. This protocol is a parametric protocol which can be used to trade-off between
delayed activities and reliability based on underlying function requirements. At different packet
loss rate, reliability and delay performance of TRM-MAC protocol is assessed by authors and
simulation investigation is used for assessing working of this protocol with other protocol.
Mohanty et al [14] recommended an energy efficient structure-free data aggregation and
delivery (ESDAD) protocol, that aggregates unnecessary data in intermediary nodes. Here,
waiting time for packets at every intermediary node is computed sensibly with intention that data
is aggregated effectively in routing path. However, sensed data packets are transferred wisely to
aggregation point for data aggregation. The ESDAD protocol calculates a cost function for
structure-free, choosing the next-hop node and executes near source data aggregation.
For WSNs-dependent smart grid applications, Faheem et al [15] presented a new Energy
Efficient and Reliable Data Gathering Routing Protocol (ODGRP). Here, this method uses a
software-described centralized organizer and numerous mobile sinks for energy effective and
reliable data collection from WSNs in SG. However, widespread experimental outcomes
performed by EstiNet 9.0 demonstrate that modelled system outperforms current technique and
attain its described aims for event-driven applications in SG.
For WSN, an interference and congestion aware energy efficient routing technique is proposed
by Elappila et al [16] and it is termed as Survivable Path Routing. In networks with high traffic,
this protocol can produce better results, where large amount of sources will try to transmit its
packets to target, which is a common condition in IoT applications for remote healthcare
monitoring. For selecting next hop node, scheme utilizes a criterion which is a task of three
features: signal to interference and link’s noise ratio, survivability factor path from subsequent
hop node to target, and congestion range at subsequent hop node.
European Journal of Molecular & Clinical Medicine ISSN 2515-8260 Volume 07, Issue 09, 2020
2884
A stable and congestion-reliant protocol was devised by Sharma et al[17], which provides
congestion control based on both bidirectional reliability and rate adjustment. Nevertheless, in
order to select the best direction for data transmission, it uses the Technique for Order Choice by
Similarity to Ideal Solution (TOPSIS) scheme, as TOPSIS prefers an alternative to have the
smallest distance from the perfect one and the highest distance from the perfect negative
resolution. Congestion is defined as a degree of congestion by using a percentage of the average
packet service time over the average packet inter-arrival time.
By changing C-SSA with Taylor series, Vinitha et al[18] resolved the energy complication and
produced an energy-efficient multi-hop routing in WSN called Taylor dependent Cat Salp
Swarm Algorithm (Taylor C-SSA). This approach passes through two stages to achieve multi-
hop routing, including Cluster Head (CH) collection and data transmission. Using the Low
Energy Adaptive Clustering Hierarchy (LEACH) protocol for the purpose of efficient data
transmission, the energy-effective CHs are calculated using the sensor nodes to transmit data
over the CH, which transfers the data to the Base Station (BS) from the best possible hop picked.
Devi et al[19] suggested the Latency and Packet Loss Reduction Cluster dependent Data
Aggregation System. Two steps are used in the formulated method: Aggregation Tree Creation
and Algorithm for Slot Scheduling. In this phase-1, each CH implements compressive
aggregation from its associates for the inward details. The aggregation tree is then built through
the sink using the Minimum Spanning Tree (MST). The packet loss rate and latency are
considered during stage-2, when prioritising and handing over time slots along with aggregated
data to the nodes. This technique avoids using redundant retransmissions and waiting, which
aims to improve the efficiency of the network.
III. PRIORITY BASED CACHE ALLOCATION AND CONGESTION CONTROL
PROTOCOL
In the proposed research work, data transmission is performed reliably by providing more
prioritization to the data flow with higher data rate. Here initially uneven cache partitioning
method is launched for giving more prioritization to the flow with higher data rates. This is
carried out by measuring the priority of the multiple data flows present in the network, and the
weight assignment will be done. Here, weight assignment is carried out through utilizing cuckoo
search algorithm where the data flow together with more priority will be assigned with more
weight. The higher cache partition will be assigned to the data flow allocated with higher weight
value. The processing flow of the proposed research method is illustrated in the figure 1 given
below.
Figure1: Proposed Work Flow
Data
packets Priority Assignment
Weight assignment
using cuckoo search
algorithm
Path capacity
calculation
Cache resource
allocation
European Journal of Molecular & Clinical Medicine ISSN 2515-8260 Volume 07, Issue 09, 2020
2885
3.1. Prioritization of Data Flow
Every packet is allocated with any one of two priorities namely dynamic or static priority in
this proposed scheme.
Static priority: It is allocated to sensor node in accordance with its data value. However,
after saving the threshold limit, sensor node is about to sense particular application for
which it is installed. The sensed value should be assessed with saved threshold limit to
allot its priority. Here, packets that carry value higher or lower than threshold limit are
noticed as high priority data packet though remaining packets are regarded as minimum
priority/ordinary packets.
Dynamic priority: It is allocated only for minimizing end-to-end delay. In a network, in
general, there are two kinds of data, namely locally created and transit or route through.
Here transit data, will have maximum priority while comparing with locally produce data
in order to minimize end-to-end delay. For accomplishing this objective, placed route
through data into priority 2 (pri2) queue at same time as locally produced data into
priority 3 (pri3) queue.
Here, the proposed priority scheme is described thoroughly in order to enhance the system
performance. While considering the first scheme, nodes with small energy construct smaller
quantity transmissions than high energy nodes, efficiently enlarging system life span. In case of
the second scheme, nodes with vital event packets have priority over nodes with data from less
vital events.
Energy-Centered Priority Scheme: The important point of this method is for amplifying
system lifetime. This is performed through minimizing transmissions count. Consequently, in
this method, data resolution is considerably minimized, that also reduces energy utilization.
While we demonstrate in results that data transmissions cannot be minimized arbitrarily, in view
of the fact that this possibly will guide to reduced system performance. Alternately, it is
suggested a fine tuning of scheme through picking proper values. On the other hand, it is obvious
that this method is not aimed for applications where major goal is to collect as much information
as feasible. Moderately, it is intended at conditions where patient possibly will be far away from
hospital or medical staff for extended time period and only common details concerning his exact
health is necessary, for example, for old people or patients suffering with chronic illness, or to
assess a brain activities between different impulses all over the day, or even for sportsmen or
women concerned in observing their physical condition. While considering the
polysomnography, this method can be helpful mainly during medical staff needs the observation
of a patient though he is residing at home and requires to examine elongated sleep phases.
We propose subsequent transmission probability for entire nodes in network. Specifically,
event description nodes along with information required to be relayed, specifically, nodes that
identified an event in the earlier frame (with probability ), send with probability explained as
where indicates transmission probability of node , represents control parameter,
indicates initial battery energy, represents residual battery energy. Here, utilizing this
European Journal of Molecular & Clinical Medicine ISSN 2515-8260 Volume 07, Issue 09, 2020
2886
transmission probability, nodes with higher residual energy broadcast with higher probability,
whereas nodes with short residual energy ranges broadcast at irregular intervals.
Data-Centered Priority Scheme. Different to the earlier method, where system life span is
enlarged by minimizing low energy nodes transmission count, in this priority method, here
aimed to accelerate highly significant packets over less significant packets, despite energy
consumption. It is to be noted that, sensors have better importance than other nodes. Here, in
order to distinguish low priority nodes from high priority nodes, we pre-assign priority range at
system setup. Furthermore, transmission probabilities are shown below:
Broadcasting probability of high priority nodes .
Broadcasting probability of low priority nodes
3.2. Weight Assignment using Cuckoo Search Algorithm
By using above procedure, priority for the nodes in wireless sensor environment would be
calculated. Priority value calculation based on, weight assignment for every node will be
performed optimally by utilizing cuckoo search algorithm. It is a kind
of optimization algorithm established by Xin-she Yang and Suash Deb in 2009 and it is
motivated by few cuckoo types obligate brood parasitism by leaving their eggs in nests of other
host birds. Few host birds can keep straight clash with interfering cuckoos. For instance, when a
host bird finds out eggs are not laid by them, it will either throw away these strange eggs away or
just discard its nest and construct a fresh nest somewhere else. Few cuckoo types, for example,
New World brood-parasitic Tapera have developed in such a manner that female parasitic
cuckoos are extremely expert in imitation in colors and model of eggs ofsome selected host
species. Cuckoo search idealize these kind of breeding activities, and as a result can be useful for
different optimization difficulties. Cuckoo Search (CS) employs following illustrations:
Every egg in nest symbolizes a solution, and cuckoo egg symbolizes novel solution. At this
point, intention is for utilizing novel and possibly better solutions (cuckoos) for changing a not-
so-fine solution in nests. Consequently, in uncomplicated way, every nest has one egg. The
scheme is stretched to make difficult cases in which every nest has several eggs demonstrating a
set of solutions. CS is dependent on three idealized rules:
1. Every cuckoo lays one egg at a particular time, and leaves its egg in an arbitrarily
selected nest;
2. Finest nests having eggs great eminencewill move to next generation;
3. Existing hosts nests count is predetermined, and egg laid by a cuckoo is found out by host
bird with a probability pa (0,1). Determining control on few set of worse nests, and
determined solutions dumped from farther calculations.
The pseudo-code can be briefly given as:
Objective function: f(x), x=(x1,x2,...,xd)
Create a primary population of n host nests;
While (t<MaxGeneration) or (stop criterion)
Obtain a cuckoo arbitrarily (say, i) and change its solution by means of carrying out Lévy
flights;
European Journal of Molecular & Clinical Medicine ISSN 2515-8260 Volume 07, Issue 09, 2020
2887
Assess its quality/fitness
[For maximization, ];
Select a nest in the midst of n (say, j) arbitrarily;
if (Fi> Fj),
Substitute j by novel solution;
end if
A fraction (pa) of inferior nests are neglected and novel ones are constructed;
Maintain most excellent solutions/nests;
Nests/solutions are ranked and recent best is computed;
Recent best solutions are passed to subsequent generation;
end while
3.3. Cache Partition Allocation
Following the distribution of data packets into their respective queues, the sensor node would
like to identify the overall path capacity. Here, end-to-end path capacity can be computed.
Moreover, the connecting two nodes, x and y point out that y indicates the parent of x in the
communication arrangement and the dotted line only demonstrates the link. To begin with the
path capacity among parent and child node can be computed. The computed path capacities are
subsequently directed to Base Station (BS). In order to compute the path capacity every node
verify radio link concerning it to its parent node. Each node transmits bulk of data packets in a
specific time to its parent node. Each data packet in the burst will be transmitted subsequent to
accepting acknowledgement of preceding data packet or time out of the preceding submitted data
packets. The path capacity Cu,v is subsequently computed through dividing the overall amount
of acknowledge packets with time taken. Subsequent to this step, every node transmits the
computed path capacity and parent node ID to BS. Then, BS carries a table that includes entries
for the entire paths form source nodes to BS. Moreover, the table include node id, id of parent
node and path capacity of the nodes in the path form source node to BS.
Subsequent to path capacities compilation at BS, further step is to compute slot period. For
the entire sensor nodes, in order to neglect congestion and to make sure the synchronization, slot
duration have to be the same during sending and receiving process for achieving
synchronization. In order to get slot duration BS select least possible path capacity of a node for
effectively transmit one data packet (Lpc). In order to transmit one data packet effectively, the
transmitting node requires 1/Lpc seconds. The achieved value will be slot period. Data loss and
congestion is subsequently neglected by means of computing an appropriate sending data rate
through BS for each sensor node u. This transmission rate possibly will not further than every
node’s sending capacity in path in direction of BS.
IV. RESULTS AND DISCUSSION
The performance assessment of proposed system P-CACC is done with RT-CacC. The
assessment was carried out in terms of parameters are known as Total Energy Consumption,
Throughput and Network Lifetime. The WSN is designed through a multiagent scheme and each
European Journal of Molecular & Clinical Medicine ISSN 2515-8260 Volume 07, Issue 09, 2020
2888
sensor is characterized by a software agent. Our multiagent scheme was generated with the NS2
simulator.The simulation of this research directs to give the optimal result than the existing
research techniques.
Total Energy Consumption: A node utilizes its energy for the purpose of sensing,
transmission and computing that energy utilization of the proposed work is low when evaluated
with RT-CacC.
Overall Network Lifetime: It can be considerably increased through sensor node’s efficient
exploitation, hat P-CACC provides much network lifetime. Here, it is high when the energy
consumption is less.
Throughput: Network throughput is described as how efficiently data is sent to sink node or
BS. It is to be pointed out that, number of data being transmitted in bits per second (bps).
Table 1: Total Energy Consumption Comparison values
Time in ms Total Energy
consumption
RT-CacC P-CACC
1 20 10
2 27 21
3 50 30
4 56 45
5 60 30
6 73 48
7 90 50
8 101 88
9 130 100
Figure 2: Total Energy Consumption at Different Time Slots
Individual node generation happens in the existing clustering methods which reasons for
more energy utilization. Total energy consumption at various time slots displayed in the above
Figure 2. Based on the assessment, it is verified that the proposed scheme P-CACC shows
30.47% lower energy consumption than RT-CACC.
0
20
40
60
80
100
120
140
1 2 3 4 5 6 7 8 9
Tota
l E
ner
gy C
onsu
mp
tio
n(j
ou
les)
Time in milliseconds
RT-CacC P-CACC
European Journal of Molecular & Clinical Medicine ISSN 2515-8260 Volume 07, Issue 09, 2020
2889
Table 2: Network Lifetime Comparison Values
Time in ms Network Lifetime
RT-CacC P-CACC
1 6000 7000
2 4500 5000
3 4000 5000
4 2000 3000
5 1800 3000
6 2000 2200
Figure 3: Overall Network Lifetime at Different Time Slots
The Network Lifetime also reduced considerably owing to much energy consumption. By
means of toggling the nodes either to active or sleep mode, energy utilization can be reduced.
Figure 3 gives the whole network lifetime at various time slots, these proposed scheme provides
more network lifetime. From this assessment, it is verified that the proposed scheme P-CaCC
shows 24.13% higher network lifetime than RT-CACC.
Table 3: Throughput Comparison Values
Time in ms Throughput
RT-CacC P-CACC
1 50000 55000
3 50000 56000
5 50000 55000
7 51000 56000
9 52000 54000
0
1000
2000
3000
4000
5000
6000
7000
8000
1 2 3 4 5 6
Net
wo
rk l
ifet
ime(
ms)
Time in milliseconds
RT-CaCC P-CACC
European Journal of Molecular & Clinical Medicine ISSN 2515-8260 Volume 07, Issue 09, 2020
2890
Figure 4: Throughput at Different Time Slots
Figure 4 demonstrates throughput at various time slots. Throughput is higher for an efficient
scheme and provides best lifetime. From this assessment, it is verified that the proposed scheme
P-CaCC shows 9.09% higher throughput than RT-CACC.
V. CONCLUSION
In the proposed research work, data transmission is executed consistently by means of
providing more prioritization to the data flow along with higher data rate. Here, initially uneven
cache partitioning method is introduced for providing more prioritization to the flow with higher
data rates. This is carried out through measuring the priority of the several data flows included in
the network, and the weight assignment will be performed. Here weight assignment is carried out
through utilizing cuckoo search algorithm where the data flow together with more priority will
be allocated with more weight. The higher cache partition will be assigned to the data flow
allocated with higher weight value. The overall evaluation of this research work is performed in
NS2 simulation, from which it is verified that proposed research work tends to have best result
than the existing schemes.
REFERENCES
1. Orhan, O. (2016). Energy neutral and low power wireless communications (Doctoral
dissertation, Polytechnic Institute of New York University).
2. Li, J., Liu, W., Wang, T., Song, H., Li, X., Liu, F., & Liu, A. (2019). Battery-friendly relay
selection scheme for prolonging the lifetimes of sensor nodes in the Internet of Things. IEEE
Access, vol. 7, pp. 33180-33201.
3. Wang, C., He, Y., Yu, F.R., Chen, Q., & Tang, L. (2017). Integration of networking, caching,
and computing in wireless systems: A survey, some research issues, and challenges. IEEE
Communications Surveys & Tutorials, vol. 20, no. 1, pp. 7-38.
4. Shaf, A., Ali, T., Draz, U., &Yasin, S. (2018). Energy Based Performance analysis of AODV
Routing Protocol under TCP and UDP Environments. EAI Endorsed Transactions on Energy
Web, vol. 5, no. 17, pp. 1-5.
47000
48000
49000
50000
51000
52000
53000
54000
55000
56000
57000
1 3 5 7 9T
hro
ugh
pu
t
Time in millisecond
RT-CaCC P-CACC
European Journal of Molecular & Clinical Medicine ISSN 2515-8260 Volume 07, Issue 09, 2020
2891
5. Yang, L., Zhu, H., Kang, K., Luo, X., Qian, H., & Yang, Y. (2018). Distributed censoring
with energy constraint in wireless sensor networks. IEEE International Conference on
Acoustics, Speech and Signal Processing (ICASSP), pp. 6428-6432.
6. Randhawa, S., & Jain, S. (2017). Data aggregation in wireless sensor networks: Previous
research, current status and future directions. Wireless Personal Communications, vol. 97, no.
3, pp. 3355-3425.
7. Tomić, I., & McCann, J. A. (2017). A survey of potential security issues in existing wireless
sensor network protocols. IEEE Internet of Things Journal, vol. 4, no. 6, pp. 1910-1923.
8. Kharb, K., Sharma, B., &Trilok, C.A. (2016). Reliable and congestion control protocols for
wireless sensor networks. International Journal of Engineering and Technology
Innovation, vol. 6, no. 1, pp. 68-78.
9. Anand, N., Varma, S., Sharma, G., &Vidalis, S. (2018). Enhanced reliable reactive routing
(ER3) protocol for multimedia applications in 3D wireless sensor networks. Multimedia
Tools and Applications, vol. 77, no. 13, pp. 16927-16946.
10. Kumaravel, K., &Anusuya, N. (2018). A Survey on Congestion Control System in Wireless
Sensor Networks. International Journal for Research in Science Engineering &
Technology, vol. 5, no. 9, pp. 7-11.
11. Harish, V.S.K.V., & Kumar, A. (2016). A review on modeling and simulation of building
energy systems. Renewable and sustainable energy reviews, vol. 56, pp. 1272-1292.
12. Rosset, V., Paulo, M.A., Cespedes, J. G., &Nascimento, M. C. (2017). Enhancing the
reliability on data delivery and energy efficiency by combining swarm intelligence and
community detection in large-scale WSNs. Expert Systems with Applications, vol. 78, pp.
89-102.
13. Bhatia, A., &Hansdah, R. C. (2016). TRM-MAC: A TDMA-based reliable multicast MAC
protocol for WSNs with flexibility to trade-off between latency and reliability. Computer
Networks, vol. 104, pp. 79-93.
14. Mohanty, P., &Kabat, M. R. (2016). Energy efficient structure-free data aggregation and
delivery in WSN. Egyptian Informatics Journal, vol. 17, no. 3, pp. 273-284.
15. Faheem, M., Butt, R. A., Raza, B., Ashraf, M. W., Ngadi, M. A., &Gungor, V. C. (2019).
Energy efficient and reliable data gathering using internet of software-defined mobile sinks
for WSNs-based smart grid applications. Computer Standards & Interfaces, vol. 66, pp. 1-18.
16. Elappila, M., Chinara, S., &Parhi, D.R. (2018). Survivable path routing in WSN for IoT
applications. Pervasive and Mobile Computing, vol. 43, pp. 49-63.
17. Sharma, B., Srivastava, G., & Lin, J. C. W. (2020). A bidirectional congestion control
transport protocol for the internet of drones. Computer Communications, vol. 153, pp. 102-
116.
18. Vinitha, A., & Rukmini, M.S.S. (2019). Secure and energy aware multi-hop routing protocol
in WSN using Taylor-based hybrid optimization algorithm. Journal of King Saud University-
Computer and Information Sciences, pp. 1-12.